1 / 19

Status of FTK & requests 2013

Old Milestones: we have even more relaxed our schedule Status of FTK work Money requests for 2013. Status of FTK & requests 2013. NEWS & Future steps TDR with tested prototypes June 2013 University of Geneva joined FTK for the LAMB project

rhoslyn
Download Presentation

Status of FTK & requests 2013

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Old Milestones: we have even more relaxed our schedule • Status of FTK work • Money requests for 2013 Status of FTK& requests 2013 • NEWS & Future steps • TDRwith tested prototypes June 2013 • University of Geneva joined FTK for the LAMB project • LPNHE and University of Tessalonikishow interest • Heidelberg got few funds (25 keuros x 4 years) • However we missmanpower ! Paola Giannetti, INFN Pisa, for the FTK Group ATLAS Italia, May 17, 2012

  2. VS: 45degree phi towers (0.3<h<1.15) 4 pix links (L1, L2) + 4 SCT links RODs per tower (0.3<h<1.15) Total 16 ROLsfor2towers 1 EDRO + 1AMB =1 tower

  3. VS idea born for m~5-10 (CDF chip, 6 layers only) – now m~30 ! • 80% bank e, m~20 • → no more than ~10 kHz event – a lot of fakes because 6 layers only • NO VS study for m~30 but certainly worse! • Take only muon triggers (use L1 ID in the header) • We will also prescale events inside FTK_IM • We will use banks with low efficiency to reduce # found roads • We will cut the Amboard output at 1000 roads. • Events will start to be complete only when m < ~20 (later in the store). • NO TRACK FITTING – ONLY ROAD FINDING! • Verify integration of FTK prototypes with ATLAS TDAQ • Verify that real-time FTK output with collision data • corresponds to FTK simulation

  4. NEW crate @CERN this week for LAB 4 Extra PC + Quest needed

  5. AMBoard generations & tests NEW:AMBFTK for 2015 (PI-MI) AMBslim5 for standard test stands AMBslim5++ for cooling tests New – 128 AMchips 6 A @ 48 Volts 270 Watts@1,8 V Old – CDF like 64 AMchips Just ordered the PCB LAMBFTK just routed (PG-PI-PV) Final crate – Wiener - soon @CERN PI-PV 5V / 345A, 3,3V /115A, 48V /81A,

  6. Next tests: AMBFTK and Proto-AUX • P3: 12 (1.5 GHz)+ 8 (3GHz) • = 20 serial links • TEST OF:- Serial links • VME on Proto AUX • Compatibility with new AMchip A lot of high frequency Serial links Backplane VME Proto AMBoard Proto AUX Milano - Pavia: Mechanics studies Tested OK ISOLA per buoncontatto P3

  7. AMCHIP04 big Jump in technology!(LNF, Pisa, Milano + ideas from USA) NEW AMCHIP Annovi, Beretta, Crescioli, Liberali, Sacco, Stabile, Hoff – Ongoing tests Consumption & perfomancesMeasured soon! 3.80 mm 65 nm 14 mm^2 = 8192 Patterns + 228 pads Variable Resolution: high fake rejection with smaller banks (Talk G. VolpiAnimma 2011) 3.31 mm 2011-12 64 patterns= 2 full custum cells

  8. What for 2015? Small production If we have to use the small chip (14 mm^2): 2barrel wedges like VS or more:: 8AMBFTKsA32 LAMBs, 1 k chips How much costs 1kchips? 500 chips →4 wafers, 7,765 k€/wafer 12 wafer=1500 chips 93 keurofor a 66%yield?? This is not too expensive for a demonstrator (small & modern) After this (2015) we can do the pilot run • 8 AMBoards → 20 keuro • 32 LAMBs → 20 keuro • 1,5 kchips → ~90 keuro ~ 130 keuro in 2014 + production FTK_IM

  9. Richieste 2013 – lineeguida • Produzionerimandataancora • 2012 sicostruiscono e testanopre-prototipi • 2013 prototipifinali e loro tests (Test stand=CDF Crates to be adjusted) • 20 k€ a Pavia per LAMB finali – Test Stand (TS) VME • 10 k€a Pisa per AMBFTK finali • 5 k€ a Milano per Test Stand (sviluppo di firmware) • 20 k€ a LNF per FTK_IM finale piccolaproduzione. • chip AM 2013: IMPORTANTE prepararsiallo step finale costoso • Crescitadell’areaper ridurreilrischio step finale (x10) • Input/output Serializzatiper liberare pads per VCC/GND • Multi-packaging of dies per aumentare la densita’ • → 100 k€a Milano per 1+ 20 k€a LNF per 2+ 20 k€ per 3 • Integration test still based on EDRO in 2013: • 5 k€ a Bologna per ‘rewarking’ of EDROs e miglioramento • test stand • SUPER TOT 200 k€

  10. Summary 2013 Milano: 5 k€ schedina chip + AMchip05, 100 k€ Pisa: 10 k€ new AMBFTK +10 k€ SJ for possible problems Pavia: 20 k€: LAMBFTK (small prod) + Test stands LNF: 20 k€ FTK_IM e 40 k€ Amchip Bologna: 5 k€ (spese di laboratorio) Tot: 200 keuro + 10 SJ Summary schedule • Produzionerimandataancora: • 2012 sicostruiscono e testanopre-prototipi • 2013 prototipifinali e lorotests • 2014 costruzionedimostratore (possibilmentecoprire Barrel) • 2015 prima presadati

  11. Backup

  12. Responsabilita’ e ME • Paola G. L2 4 mesi • Marco P. L2 4 mesi • Agostino L. L2 1 mese • Andrea N. L2 1,5 mesi • Annovi A. L2 4 mesi • Volpi G. L3 2 mesi • Mauro V. L3 3 mesi • A. Stabile L3 1 mese • Valentino L. L3 1 mese Responsabilita’ descritte a questa pagina:https://twiki.cern.ch/twiki/bin/viewauth/Atlas/FastTracKer Integrazione di FTK con prototipiamericani @CERN: Pisa 1 mesi Milano 1 mesi Frascati 1 mesi Pavia 1mesi Bologna 1 mesi

  13. Missioni interne 4 K€ Frascatiintegrazioneprototipi e AMchip I e II 5 K€ Pisa integrazioneprototipi 4 K€ Milano AMchip I e II 4 K€ Pavia integrazioneprototipi Bologna integrazioneprototipi

  14. Conclusions • We are working hard for both the Final FTK • & the intermediate versions (vertical slice and demonstrator) • First prototypes (chip and boards) will be under tests in June. • the AMchip just arrived, as planned. • the vertical slice plan has been reduced (LHC schedule). • the TDR in 2013 should start final approvals • A demonstrator expected in 2015, big production will start after.

  15. ANALYSIS of VS data • This task is independent of the hardware task: • If hardware produces the same roads as FTKsim → • can do everything else with FTKsim – • F. Crescioli since February: try to reconstruct top events in VS: • apply track isolation to leptons • measure rejection for muons from b-jets • measure efficiency for muons from Ws • In 3 months: (a) choosing the right strategy to have FTK tracks available to trigger algorithms, (b) learning to run muComb – muIso on them. • V. Cavaliereand A. McCarn are joining Francesco’s effort now. Additional help would be welcomed from HLT experts. • Fakes will be a problem – but let’s prepare the algos

  16. DONE 2013 A minimum of Two 45 degree wedges in the barrel • If the run is extended to 2013, enlarge the vertical slice • in 2012 for real triggering → “small demonstrator” • primary vertex identification, beam spot calculation ……. • Track-isolated muons, highly ionizing particles ….. NO run EXTENSION

  17. FTK: Projects to Institutions Frascati-Waseda: Clustering in Pixels-SCT mezznines Pisa & Frascati & Fermilab the AMChip could be 3D USA-Italy “2-outHola” Pisa: AMBoard Illinois: Final board in core crate Argonne Raw data ROBs Chicago GigaFitter + Data Organizer + HW Pixels & SCT RODs 8 “core” crates: AMboard Data Organizer GigaFitter 4 DF crates: cluster finding split by layer Fermilab: DF motherboards S-links 1 Readout Crate Track data ROB USA+Italy: crates + links + backplanes…

  18. OPTION A CPU vme AM10+….. AM15+….. AM14+….. AM13+…… AM9+…... AM12+….. AM11+….. AM8+….. AM7+TSP+DO+TF+HW AM6+TSP+DO+TF+HW AM4+TSP+DO+TF+HW AM5+TSP+DO+TF+HW AM0+TSP+DO+TF+HW AM3+TSP+DO+TF+HW AM2+TSP+DO+TF+HW AM1+TSP+DO+TF+HW 11LayFit+ HW final 11LayFit+ HW final 11LayFit+HW 11LayFit+HW DO+TF+HW Standard cell chip Interface Connectors for tracks output LAMB DO TF HW Control FPGA FPGA for Roads TF HW DO TF HW DO Connectors for Hits LVDS Cables 40 MHz clock TF HW DO FPGA for SS Input P3 serial LVDS DO INPUT FIFOs SSMAP AMBoard ITALIAN DUTIES for 3x10**34 PISA DATA FORMATTER FRASCATI Processing Unit Pisa +Chicago AUX card FERMILAB

More Related